Infra as Code
In this demo, a complete infrastructure to deploy your application is created in matter of minute with a single command line execution via Terraform.
See the demo SourceLLM Chatbot
A LLM Chatbot leveraging RAG with embedding is running on our newest Nvidia L40s with 48 GB of video RAM and the pgvector extension of our managed PostgreSQL database.
See the demo SourceObject detection
Using the Nvidia L4 with 24 GB of video RAM for the astounding price of 0.75 Euros per hour, this demo shows object detection in pictures using the Ultralytics YOLOv8 model.
See the demo SourceAmpere inference
Using ARM optimized framework and leveraging Scaleway Cost Optimized ARM instance, this demonstrate the ability to execute Speech recognition using Whisper inference on Ampere infrastructure.
See the demo Source